Why two photos the same size in pixels have very different file sizes

Two photos with identical pixel dimensions can differ several times in file size because compression depends on content. Smooth skies and plain walls compress well; grass, gravel, text and noise from low light do not. Format, quality setting and embedded data such as thumbnails and location add to the difference.

Pixels set the ceiling, content sets the size

An uncompressed 4000 × 3000 photo would be about 36 MB: 12 million pixels, three bytes each. Formats such as JPG, WebP and HEIC get that down by describing the image more cleverly — “this whole patch is roughly the same blue” is shorter than listing every pixel. How much they can save depends on how predictable the picture is.

• Low detail compresses well: sky, a blurred background, a white wall, a document with lots of white paper.
• Fine detail does not: foliage, gravel, fabric, crowds, small text.
• Noise is the hidden culprit. Photos taken in dim light carry random speckle, and random detail is the hardest thing to compress. The same room photographed at night can be much larger than by day.

Other things that change the size

• Quality setting. JPG quality 95 can be two to three times the size of quality 80, with very little visible difference.
• Format. PNG is lossless and is several times larger than JPG for photos. HEIC and WebP are typically smaller than JPG at similar quality.
• Metadata. Camera details, GPS location, a preview thumbnail and sometimes depth or editing data ride along inside the file. Usually this is tens of kilobytes, occasionally more.
• Editing history. A photo that was edited and re-saved, or passed through several apps, may have been re-encoded at a different quality each time.

See it for yourself

Take two photos on the same phone, one after the other: one of a plain wall or the sky, one of a bush or a bookshelf. Then compare the file sizes. Same camera, same settings, same pixel count — and often a large difference in size, because the bush gives the encoder far more to describe. Then take the same bookshelf shot again with the lights dimmed. The extra noise from low light usually pushes the size up again, even though the picture looks darker and “emptier”. It is a quick way to see that file size measures how much information there is to store, not how big the picture is.

Why it matters for upload limits

Forms that ask for a photo “under 200 KB” do not care about pixels, only bytes. This is why advice like “resize to 1200 pixels” works for one photo and fails for the next. A busy outdoor shot at 1200 pixels may still be over the limit when a portrait against a plain wall is well under.

How to shrink a big one

1. Aim at the number, not the pixels. Compress Image can target a file size such as 200 KB and work out the quality for you, resizing if needed. It runs in your browser.
2. Strip what you do not need. EXIF Remover cuts out location, camera data and embedded previews without re-compressing the image — a modest saving, and a privacy win.
3. Switch format if allowed. If the destination accepts WebP, the Image Converter will usually produce a smaller file than JPG at similar quality. Never save a photo as PNG to make it smaller.
4. Crop away busy areas. Cropping out grass, gravel or a cluttered background removes the most expensive pixels first.

Mistakes to avoid

• Assuming a smaller file means a worse photo; a plain subject is simply cheaper to store.
• Running a denoise filter and expecting a small file — some help, many re-save at high quality.
• Converting to PNG to “keep quality” before an upload with a size limit.

Checklist

• Target is in KB or MB? Compress to the size, not the pixels.
• Metadata removed if location or device details are not needed?
• Format suited to photos: JPG, WebP or HEIC, not PNG?